Foundations, Applications & Theory of Inductive Logic (FAT IL)

Breadcrumb Navigation



Martin Adamčík (Assumption University, Thailand) is an Assistant Professor in Mathematics at the Assumption University of Thailand. Before joining the Assumption University, he obtained a Ph.D. degree in Mathematical Logic at the University of Manchester supported by a Marie Curie Early Stage Fellowship. Recently, he has developed a method for meta analysis that can deal with both complex knowledge and unexplained heterogeneity [2017]. This method was argued for using a Bayesian argument [2016] similar in spirit to the one used to justify the maximum entropy inference process. Currently, he’s interested in objective Bayesianism and developing a practical program for meta analysis directly usable by medical researchers. For more information, please visit his website.

Hykel Hosni (Milan) is Associate Professor of Logic at the Department of Philosophy of the University of Milan. He is also Adjunct Professor of Logic and methodology of the social sciences at Bocconi University, Milan. He works on the logical foundations of uncertain reasoning and its applications to decision theory. The results of his research have appeared in leading logic journals, including the Journal of Symbolic Logic, and on leading AI journals, including the International Journal for Approximated Reasoning. His most recent research investigated the foundations and methodology of forecasting based on “big data” and its applicability to the diagnosis of rare cancers. He holds a BA and an MA in Philosophy (Pisa) and a PhD in Mathematics (Manchester). He is also author of a popular book on probability published in Italian in 2018 (Probabilit'a: come smettere di preoccuparsi e imparare ad amare l’incertezza.) For more information, please visit his website.

Gabriele Kern-Isberner (TU Dortmund) is a Professor for Information Engineering at the department of computer science at the Technische Universitaet Dortmund. Her scientific work focuses on qualitative and quantitative approaches to knowledge representation, such as description logics and ontologies, default and non-monotonic logics, uncertain reasoning, probabilistic reasoning, belief revision, and argumentation, and she is also interested in multi-agent systems and knowledge discovery. Her research works include in particular the development of methods that help integrate approaches and points of views from different fields, such as the combination of first-order logic and probabilities, or building bridges between probabilistic and nonmonotonic reasoning. Some of her works also deal with the cognitive evaluation of formal models of reasoning and the development of suitable formal models for rational human reasoning. She has been working closely together with cognitive scientists and philosophers in the DFG Priority Programme ”New Frameworks of Rationality” (DFG SPP 1516). She holds a doctoral degree in mathematics and completed her habilitation in computer science in 2000. For more information, please visit her website.

Dominik Klein (Bayreuth/Bamberg) is a postdoctoral researcher at the Universities of Bayreuth and Bamberg in Germany (2015-2021). He holds a PhD in philosophy from Tilburg University (2015) and a Masters (German: Diplom) in Mathematics from the University of Bonn. He has worked on Multi- Agent Logic and on the relationship between Logic and Probability Theory. In the latter area, he has established contacts between the two fields in both directions. In one way of contact, he has applied logic to classically probabilistic domains such as mixed strategies in games. Conversely, he has employed quantitative methods, discrete time dynamic systems, for the study of dynamic logics. Jointly with Soroush Rafiee Rad, he has also enquired into probabilistic generalisations of Belnap-Dunn four-valued logics. For more information, please visit his website.

Jürgen Landes (LMU Munich) is based at the Munich Center for Mathematical Philosophy (MCMP) where he is the PI of the DFG funded project Evidence and Objective Bayesian Epistemology (2018- 2021). He holds a PhD in mathematics (Manchester University, co-supervised by Vencovská) for his work on polyadic inductive logic. In between, he held four postdoc positions concerned with uncertainty: multicriterial decision making for the environmental impacts of micro algae which produce methane for energy production (2009-2010), automated contract negotiations in a temporary employment agency (2010-2012), objective Bayesian epistemology and inductive logic (2012-2015, PI Williamson) and Bayesian approaches to the philosophy of pharmacology and evidence (2015-2018). He has worked and published on Carnapian inductive logic, MaxEnt, multi-criterial decision making, multi-agent systems, general philosophy of science, philosophy of medicine and quantum information theory. His relevant current research interests are in objective Bayesian inductive logic: theory (predicate languages) and application via Bayesian networks (evidence aggregation and computational approaches). He currently cooperates with Rafiee Rad and Jon Williamson on inductive logic (2 joint manuscripts), frequently contributes to The Reasoner edited by Hosni and founded by Williamson; his interview with Kern-Isberner is in this February’s issue of The Reasoner [2019]. In the past, he guest-edited papers by Adamčík and Kern-Isberner (Beierle et al. [2015]) and was a collaborator of Vencovská. For more information, please visit his website.

Soroush Rafiee Rad (Bayreuth) is a postdoctoral researcher at Bayreuth University working on a project in collective attitude formation (2017-2020). Previously, he was a postdoc at the Institute for Logic, Language and Computation at the University of Amsterdam and a research fellow at Munich Center for Mathematical Philosophy. He completed two PhDs – in Philosophy at Tilburg Center for Logic and Philosophy of Science and in Mathematics at the University of Manchester. His research lies at the intersection of mathematical logic, probabilistic reasoning and formal epistemology. He has worked on the probabilistic inference processes on first order languages, maximum entropy and probabilistic characterisations of models of first order theories, rational belief formation, doxastic logics for learning probabilities and probabilistic extension of Bell-Dunn logic (with Dominik Klein) as well as models of rational deliberation and preference aggregation. For more information, please visit his website.

Jan-Willem Romeijn (Groningen) is professor of philosophy of science and head of the department of theoretical philosophy in the Faculty of Philosophy of the University of Groningen. His research interests includ statistical inference, formal and social epistemology, and general philosophy of science. He has collaborated with a wide variety of disciplines in the humanities and the exact, social and medical sciences, and he provides advice on uncertain reasoning and social deliberation in judicial and policy contexts. He has recently submitted a research proposal on the intersection of inductive logic and the philosophy of data science. For more information, please visit his website.

Tom Sterkenburg (LMU Munich) is currently (2017-2019) a Postdoctoral Fellow at the Munich Center for Mathematical Philosophy, where he studies the philosophical foundations of machine learning. He holds a BSc in Artificial Intelligence (Amsterdam), a MSc in Logic (Amsterdam), a MSc in History and Philosophy of Science (Utrecht), and a PhD in Philosophy (joint at the University of Groningen and the CWI, the Dutch national research center for mathematics and computer science); all degrees were awarded cum laude. In his PhD thesis (2018), for which he received the Wolfgang Stegmüller Award, he investigated the computability-theoretic approach to probabilistic “universal prediction” as a link between Carnap’s inductive logic and modern approaches in machine learning. For more information, please visit his website.

Marta Sznajder (LMU Munich) is a Postdoctoral Fellow at the Munich Center for Mathematical Philosophy. Previously she held a postdoctoral position at the Czech Academy of Sciences and was a researcher at the University of Groningen. Her research concerns the history and philosophical context of Carnapian inductive logic. She is particularly interested in the normative import of inductive logic on belief, and its relation to practical rationality. Her doctoral project was concerned with the relation between inductive logic and the theory of conceptual spaces. In 2018, she was awarded the Kristeller-Popkin Travel Fellowship for research in the Carnap archives in Pittsburgh and Los Angeles. She holds a BA and MA in Philosophy (Wrocław), an MSc in Logic (Amsterdam), and a PhD in Philosophy from LMU Munich and the University of Groningen (double degree). Within the network, she is looking forward to enriching her historical work with the insights learned from the practicioners of inductive logic coming from the sciences. For more information, please visit her website.

Paul Thorn (Düsseldorf) is a postdoctoral researcher at Heinrich Heine University Düsseldorf (2009-2019), currently working as a member of the Collaborative Research Center: The Structure of Representations in Language, Cognition, and Science (DFG SFP 991). Paul works in a number of areas of formal epistemology, two of which are directly relevant to the research network. First, Paul has worked extensively on the topic of direct inference (i.e., inference to single-case probabilities from frequency information). Past and continuing work have aimed at clarifying (1) the correct rules of direct inference, and (2) the connection between direct inference and the principle of indifference, and by extension entropy based probability logic. Second, Paul has conducted simulation-based studies of numerous inference rules and systems of inference. Previous work evaluated meta-inductive inference and a number of formal systems with probabilistic and/or ranking based semantics. Current work is focussed on the connections between inheritance inference and cluster analysis. A reoccurring focus of the simulation studies has been dependencies in the accuracy of inference (of various sorts) on the entropy level of the environment. For more information, please visit his website.

Matthias Thimm (Koblenz-Landau) is a senior researcher at the Institute for Web Science and Technologies (WeST) in Koblenz, Germany (current contract is until January 2020). He received his PhD degree from the university of Dortmund (Germany) in 2011 and his habilitation degree from the university of Koblenz-Landau (Germany) in 2016. His research focus is on formal methods of knowledge representation and artificial intelligence, both from a conceptual as well as algorithmic perspective. He is interested in formal models of argumentation, in particular with respect to quantitative extensions, game theoretical aspects for application in multi-agent systems, the relationship of argumentation and belief revision, and inconsistency measurement. Particularly relevant for this project are previous works on probabilistic reasoning with incomplete and inconsistent information in propositional and first-order representations of knowledge. These works deals with the application of the principle of maximum entropy in scenarios of rational decision-making in artificial intelligence, most importantly in scenarios where contradicting pieces of information have to be resolved in order to guarantee a robust working system. For more information, please visit his website.

Alena Vencovská (Manchester) has been a researcher on fixed terms contracts at the University of Manchester since 1984 (with various breaks). Currently she is a visitor at this institution, with a part-time position at the Open University. Her research interests include the foundations and history of mathematics and uncertain reasoning, in particular pure inductive logic. Results of her work, mostly joint with Jeff Paris, have appeared in high ranking journals like Journal of Symbolic Logic, Erkenntnis, Synthese, Journal of Philosophical Logic and Annals of Pure and Applied Logic. For more information, please visit her website.

Jon Williamson (Kent) is Professor of Reasoning, Inference and Scientific Method in the Philosophy Department and co-director of the Centre for Reasoning. Jon works on inductive logic and the philosophy of science and medicine. With respect to inductive logic, he has worked on a unifying framework for probabilistic logic (Probabilistic logics and probabilistic networks, Springer 2011) and developed a Bayesian version of inductive logic (Lectures on inductive logic, OUP 2017). He has also investigated the use of evidence of mechanisms in medicine (see Evaluating evidence of mechanisms in medicine, Springer 2018, and the EBM+ network), developed an epistemic theory of causality (Bayesian nets and causality, OUP 2005), and defended an analogous epistemic theory of probability (In defence of objective Bayesianism, OUP 2010). He holds a BA in Mathematics (Manchester), an MA in Philosophy of Science and Mathematics and a PhD in Philosophy (both King’s College, London). For more information, please visit his website.

Francesca Zaffora Blando (Stanford) is a PhD candidate in Philosophy and Symbolic Systems at Stanford University, where she is a recipient of the Patrick Suppes Fellowship in Philosophy of Science. She also holds an MA in Philosophy (Edinburgh) and an MSc in Logic (ILLC, Amsterdam). Her PhD dissertation, supervised by Johan van Benthem (Stanford, Amsterdam, Tsinghua), centres around the interconnections between algorithmic randomness—a branch of computability theory—and various formal models of learning, including formal learning theory and Bayesian epistemology. In a recent paper, A learning-theoretic characterisation of Martin-L¨of randomness and Schnorr randomness (currently under review), she uses notions from formal learning theory to provide novel characterisations of standard algorithmic randomness notions in terms of unlearnability. Her most recent project, in collaboration with Simon Huttegger (UC Irvine) and Sean Walsh (UCLA), focuses on applying algorithmic randomness to classical Bayesian convergence-to-the-truth [Huttegger, 2015] and merging-of-opinions results [Blackwell and Dubins, 1962, Gaifman and Snir, 1982]. The authors consider computable versions of these classical results and provide precise characterisations of the collection of environments in which computable Bayesian agents are inductively successful using algorithmic randomness notions. More specifically, they show that the random worlds, or environments, are exactly the ones in which a certain kind of inductive success is attainable. For more information, please visit her website.